Noise Robust Human Detection Combining Self-Quotient ε-Filter and HOG Feature Distance Criterion

Mitsuharu Matsumoto


Human detection, Self-quotient ε-filter, Histograms of oriented gradients, Parameter setting


This paper describes human detection using self-quotient ε-filter (SQEF) with the parameter determined by Histograms of Oriented Gradients (HOG) feature distance criterion. Although human detection combining HOG and SVM is a powerful approach, as it uses local intensity gradients, it is difficult to handle noise corrupted images. On the other hand, although human detection combining SQEF, HOG and SVM can realize noise robust human detection, SQEF requires manual parameter setting. Our aim is to realize noise robust human detection by using HOG feature distance criterion. Our aim is not only to set the parameter of self-quotient ε-filter but also to train SVM by using numerous images without noise and a small amount of images with noise.

Important Links:

Go Back